Wireless communication systems are well known in the art. Communications standards are developed in order to provide global connectivity for wireless systems and to achieve performance goals in terms of, for example, throughput, latency and coverage. One current standard in widespread use, called Universal Mobile Telecommunications Systems (UMTS), was developed as part of Third Generation (3G) Radio Systems, and is maintained by the Third Generation Partnership Project (3 GPP).
A typical UMTS system architecture in accordance with current 3 GPP specifications is depicted in FIG. 1. The UMTS network architecture includes a Core Network (CN) interconnected with a UMTS Terrestrial Radio Access Network (UTRAN) via an Iu interface. The UTRAN is configured to provide wireless telecommunication services to users through wireless transmit receive units (WTRUs), referred to as user equipments (UEs) in the 3 GPP standard, via a Uu radio interface.
For example, a commonly employed air interface defined in the UMTS standard is wideband code division multiple access (W-CDMA). The UTRAN has one or more radio network controllers (RNCs) and base stations, referred to as Node Bs by 3 GPP, which collectively provide for the geographic coverage for wireless communications with UEs. One or more Node Bs are connected to each RNC via an Iub interface. RNCs within a UTRAN communicate via an Iur interface.
The Uu radio interface of a 3 GPP system uses Transport Channels (TrCHs) for transfer of higher layer packet containing user data and signaling between UEs and Node Bs. In 3 GPP communications, TrCH data is conveyed by one or more physical channels defined by mutually exclusive physical radio resources, or shared physical radio resources in the case of shared channels.
To improve reliability of data transmission, automatic repeat request (ARQ) or hybrid ARQ (HARQ) is implemented. HARQ and ARQ employ a mechanism to send feedback to the original sender in the form of a positive acknowledgment (ACK) or a negative acknowledgement (NACK) that respectively indicate successful or unsuccessful receipt of a data packet to a transmitter so that the transmitter may retransmit a failed packet. HARQ also uses error correcting codes, such as turbo codes, for added reliability.
Evolved universal terrestrial radio access (E-UTRA) and UTRAN long term evolution (LTE) are part of a current effort lead by 3 GPP towards achieving high data-rate, low-latency, packet-optimized system capacity and coverage in UMTS systems. In this regard, LTE is being designed with significant changes to existing 3 GPP radio interface and radio network architecture, requiring evolved Node Bs (eNBs), which are base stations (Node Bs) configured for LTE. For example, it has been proposed for LTE to replace code division multiple access (CDMA) channel access used currently in UMTS, by orthogonal frequency division multiple access (OFDMA) and frequency division multiple access (FDMA) as air interface technologies for downlink and uplink transmissions, respectively. LTE is being designed to use HARQ with one HARQ process assigned to each data flow and include physical layer support for multiple-input multiple-output (MIMO).
LTE systems are also being designed to be entirely packet switched for both voice and data traffic. This leads to many challenges in the design of LTE systems to support voice over internet protocol (VoIP) service, which is not supported in current UMTS systems. VoIP applications provide continuous voice data traffic such that data rates vary over time due to intermittent voice activity. Variable data rate applications like VoIP provide specific challenges for physical resource allocation, as described below.
eNBs in LTE are responsible for physical radio resource assignment for both uplink (UL) communications from a UE to the eNB, and downlink (DL) communications from eNB to a UE. Radio resource allocation in LTE systems involves the assignment of frequency-time (FT) resources in an UL or DL for a particular data flow. Specifically, according to current LTE proposals, FT resources are allocated according to blocks of frequency subcarriers or subchannels in one or more timeslots, generally referred to as radio blocks. The amount of physical resources assigned to a data flow, for example a number of radio blocks, is typically chosen to support the required data rate of the application or possibly other quality of service (QoS) requirements such as priority.
It has been proposed that physical resource allocation for DL and UL communications over the E-UTRA air interface in LTE can be made valid for either a predetermined duration of time, known as non-persistent assignment, or an undetermined duration of time, known as persistent assignment. Since the assignment messages transmitted by the eNB may target both the intended recipient UE of the assignment as well as any UEs currently assigned to the resources specified by the assignment, the eNB may multicast the assignment message, such that the control channel structure allows for UEs to decode control channel messages targeted for other UEs.
For applications that require sporadic resources, such as hypertext transport protocol (HTTP) web browser traffic, the physical resources are best utilized if they are assigned on an as-need basis. In this case, the resources are explicitly assigned and signaled by the layer 1 (L1) control channel, where L1 includes the physical (PHY) layer. For applications requiring periodic or continuous allocation of resources, such as for VoIP, periodic or continuous signaling of assigned physical resources may be avoided using persistent allocation. According to persistent allocation, radio resource assignments are valid as long as an explicit deallocation is not made. The objective of persistent scheduling is to reduce L1 and layer 2 (L2) control channel overhead, especially for VoIP traffic, where L2 includes the medium access control (MAC) layer. Persistent and non-persistent assignments by the L1 control channel may be supported using, for example, a persistent flag or a message ID to distinguish between the two types of assignment in an assignment message transmitted by the eNB.
FIGS. 2 and 3 illustrate examples of persistent allocation of frequency-time resources in LTE, where each physical layer sub-frame comprises four time interlaces to support HARQ retransmissions of negatively acknowledged data. Each interlace is used for the transmission of a particular higher layer data flow, such that the same interlace in a subsequent sub-frame is used for retransmission of packets that were unsuccessfully transmitted. A fixed set of frequency-time (FT) resources are assigned in each interlace for control traffic as a control channel, which may include the L1 common control channel (CCCH) and synchronization channel.
FIG. 2 shows an example of persistent allocation and deallocation. In sub-frame 1, a first set of frequency-time resources (FT1), including one or more radio blocks, are allocated to UE1 via the control channel. Assuming the transmission of data to UE1 completes after i-1 sub-frames, the eNB sends in sub-frame i a control message to UE1 and UE2 in order to deallocate resources FT1 from UE1 and allocate them to UE2. The control channel can be used in the intermediate sub-frames between sub-frames 1 and i for the assignment of other FT resources. FIG. 3 shows an example of persistent allocation and expansion, where eNB assigns additional physical resources FT2 to UE1 in sub-frame i to support additional traffic for UE1.
A characteristic of many real time services (RTS), such as voice services, is variable data rates. In the case of voice services, a conversation is characterized by periods of speech followed by periods of silence, thus requiring alternating, constantly varying data rates. For example, a typical adaptive multi-rate (AMR) channel for voice service supports eight encoding rates from 4.75 Kbps to 12.2 Kbps and a typical adaptive multi-rate wide-band (AMR-WB) channel supports nine encoding rates from 6.6 Kbps to 23.85 Kbps.
Current techniques for persistent resource scheduling are not designed to accommodate variations in data rates. Under conventional persistent allocation, physical resources are allocated to support either a maximum data rate for a data flow or some sufficiently large fixed data rate supported by the physical channel. Accordingly, physical resources are wasted because the resource allocation is not able to adapt to changes in required data rates resulting from, for example, intermittent voice activity.
In order to support variable data rates, an eNB must be signaled the changing data rates for both UL and DL traffic. In LTE systems, an eNB can easily monitor DL data rate variations that originate at the eNB and make efficient DL resource assignment. However, current UMTS systems and proposals for LTE systems do not provide a manner for an eNB to monitor data rate variations for UL traffic originating at a UE so that the eNB may accordingly assign the appropriate amount of UL physical resources in a dynamic and efficient manner. Additionally, current proposals for LTE systems do not support high-level configuration operations for VoIP service.
The inventors have recognized that there is a need in LTE systems for support of dynamic resource allocation in combination with persistent resource allocation, along with efficient scheduling and control signaling, in order to support RTS applications with changing data rates such as VoIP. Therefore, the inventors have developed a method and apparatus for solving these problems in LTE systems.